Automating aggressive value intelligence with Amazon Nova Act

Monitoring competitor costs is crucial for ecommerce groups to keep up a market edge. Nevertheless, many groups stay trapped in guide monitoring, losing hours every day checking particular person web sites. This inefficient method delays decision-making, raises operational prices, and dangers human errors that lead to missed income and misplaced alternatives.

Amazon Nova Act is an open-source browser automation SDK used to construct clever brokers that may navigate web sites and extract information utilizing pure language directions. This put up demonstrates the way to construct an automatic aggressive value intelligence system that streamlines guide workflows, supporting groups to make data-driven pricing choices with real-time market insights.

The hidden value of guide aggressive value intelligence

Ecommerce groups want well timed and correct market information to remain aggressive. Conventional workflows are guide and error-prone, involving looking a number of competitor web sites for sure merchandise, recording pricing and promotional information, and consolidating this information into spreadsheets for evaluation. This course of presents a number of crucial challenges:

Time and useful resource consumption: Handbook value monitoring consumes hours of workers time day by day, representing a big operational value that scales poorly as product catalogs develop.
Information high quality points: Handbook information entry introduces inconsistency and human error, probably resulting in incorrect pricing choices based mostly on flawed data.
Scalability limitations: As product catalogs develop, guide processes develop into more and more unsustainable, creating bottlenecks in aggressive evaluation.
Delayed insights: Essentially the most crucial difficulty is timing. Competitor pricing can change quickly all through the day, which means choices made on stale information may end up in misplaced income or missed alternatives.

These challenges lengthen far past ecommerce. Insurance coverage suppliers routinely assessment competitor insurance policies, inclusions, exclusions, and premium constructions to keep up market competitiveness. Monetary companies establishments analyze mortgage charges, bank card provides, and price constructions via time-consuming guide checks. Journey and hospitality companies monitor fluctuating costs for flights, lodging, and packages to regulate their choices dynamically. Whatever the trade, the identical struggles exist. Handbook analysis is sluggish, labor-intensive, and susceptible to human error. In markets the place costs change by the hour, these delays make it virtually inconceivable to remain aggressive.

Automating with Amazon Nova Act

Amazon Nova Act is an AWS service, with an accompanying SDK, designed to assist builders construct brokers that may act inside net browsers. Builders construction their automations by composing smaller, focused instructions in Python, combining pure language directions for browser interactions with programmatic logic resembling assessments, breakpoints, assertions, or thread-pooling for parallelization. By its instrument calling functionality, builders may also allow API calls alongside browser actions. This provides groups full management over how their automations run and scale. Nova Act helps agentic commerce situations the place automated brokers deal with duties resembling aggressive monitoring, content material validation, catalogue updates, and multi-step looking workflows. Aggressive value intelligence is a powerful match as a result of the SDK is designed to deal with real-world web site habits, together with format adjustments and dynamic content material.

Ecommerce websites often change layouts, run short-lived promotions, or rotate banners and parts. These shifts usually break conventional rules-based scripts that depend on fastened ingredient selectors or inflexible navigation paths. Nova Act’s versatile, pure language command-driven method helps brokers proceed working at the same time as pages evolve, offering the resilience wanted for manufacturing aggressive intelligence programs.

Frequent constructing blocks

Nova Act features a set of constructing blocks that simplify browser automation. This can be utilized by ecommerce firms to gather and report product costs from web sites with out human intervention. The constructing blocks that allow this embody:

Extracting data from a webpage

With the extraction capabilities in Nova Act, brokers can collect structured information instantly from a rendered webpage. You possibly can outline a Pydantic mannequin that represents the schema that they need returned, then ask an act_get() name to reply a query concerning the present browser web page utilizing that schema. This retains the extracted information strongly typed, validated, and prepared for downstream use.

Nova.act_get("Seek for 'iPad Professional 13-inch (M4 chip), 256GB Wi-Fi'.", schema=ProductData.model_json_schema())

Navigate to a webpage

This step redirects the agent to a selected webpage as a place to begin. A brand new browser session opens at a desired start line, enabling the agent to take actions or extract information.

nova.go_to_url(website_url)

Operating a number of classes in parallel

Value intelligence workloads usually require checking dozens of competitor pages in a brief interval. A single Nova Act occasion can invoke just one browser at a time, however a number of cases can run concurrently. Every occasion is light-weight, making it sensible to spin up a number of in parallel and distribute work throughout them. This allows a map‑cut back fashion method to browser automation the place completely different Nova Act cases deal with separate duties on the identical time. By parallelizing searches or extraction work throughout many cases, organizations can cut back complete execution time and monitor giant product catalogs with minimal latency.

from concurrent.futures import ThreadPoolExecutor, as_completed

from nova_act import ActError, NovaAct

# Accumulate the entire checklist right here.
all_prices = []

# Set max staff to the max variety of lively browser classes.
with ThreadPoolExecutor(max_workers=10) as executor:
    # Get all costs in parallel.
    future_to_source = {
        executor.submit(
            check_source_price, product_name, source_name, source_url, headless
        ): source_name
        for source_name, source_url in sources
    }
    # Acquire the ends in all_books.
    for future in as_completed(future_to_source.keys()):
        attempt:
            supply = future_to_source[future]
            source_price = future.end result()
            if source_price isn't None:
                all_prices.lengthen(source_price.supply)
        besides ActError as exc:
            print(f"Skipping supply value on account of error: {exc}")

print(f"Discovered {len(all_prices)} supply costs:n{all_books}")

Captchas

Some web sites current captchas throughout automated looking. For moral causes, we suggest involving a human to unravel captchas fairly than making an attempt automated options. Nova Act doesn’t remedy captchas on the person’s behalf.

When working Nova Act regionally, your workflow can use an act_get() name to detect whether or not a captcha is current. If one is detected, the workflow can pause and immediate the person to finish it manually, for instance, by calling enter() in a terminal-launched course of. To allow this, run your workflow in headed mode (set headless=False, which is the default) so the person can work together with the browser window instantly.

When deploying Nova Act workflows with AgentCore Browser Device (ACBT), you should use its built-in human-in-the-loop (HITL) capabilities. ACBT gives serverless browser infrastructure with stay streaming from the AgentCore AWS Console. When a captcha is encountered, a human operator can take over the browser session in real-time via the UI takeover characteristic, remedy the problem, and return management to the Nova Act workflow.

end result = nova.act("Is there a captcha on the display?", schema=BOOL_SCHEMA) if end result.matches_schema and end result.parsed_response:
    enter("Please remedy the captcha and hit return when accomplished")
...

Dealing with errors

As soon as the Nova Act consumer is began, it might encounter errors throughout an act() name. These points can come up from dynamic layouts, lacking parts, or sudden web page adjustments. Nova Act surfaces these conditions as ActErrors in order that builders can catch them, retry operations, apply fallback logic, or log particulars for additional evaluation. This helps value intelligence brokers keep away from silent failures and proceed working even when web sites behave unpredictably.

Constructing and Monitoring Nova Act workflows

Constructing with AI-powered IDEs

Builders constructing Nova Act automation workflows can speed up experimentation and prototyping by utilizing AI-powered improvement environments with Nova Act IDE extensions. The extension is out there for in style IDEs together with Kiro, Visible Studio Code, and Cursor, bringing clever code era and context-aware help instantly into your most popular improvement atmosphere. The IDE extension for Amazon Nova Act accelerates improvement by turning pure language prompts into production-ready code. As an alternative of digging via documentation or writing repetitive boilerplate, you possibly can merely describe your automation objectives. That is useful for complicated duties like aggressive value intelligence, the place the extension may help you rapidly construction ThreadPoolExecutor logic, design Pydantic schemas, and construct strong error dealing with.

Observing workflows within the Nova Act console

The Nova Act AWS console gives visibility into your workflow execution with detailed traces and artifacts out of your AWS atmosphere by way of the AWS Administration Console. It gives a central place to handle and monitor automation workflows in real-time. You possibly can navigate from a high-level view of the workflow runs into the particular particulars of particular person classes, acts, and steps. This visibility lets you debug and analyze efficiency by displaying you precisely how the agent makes choices and executes loops. With direct entry to screenshots, logs, and information saved in Amazon S3, you possibly can troubleshoot points rapidly with out switching between completely different instruments. This streamlines the troubleshooting course of and accelerates the iteration cycle from experimentation to manufacturing deployment.

Operating the answer

That will help you get began with automated market analysis, we’ve launched a Python-based pattern mission that handles the heavy lifting of value monitoring. This answer makes use of Amazon Nova Act to launch a number of browser classes directly, looking for merchandise throughout varied competitor websites concurrently. As an alternative of going via tabs your self, the script navigates the online to search out costs and promotions. It then gathers every little thing right into a clear, structured format so you should use it in your personal pricing fashions. The next sections will describe how one can get began constructing the aggressive value intelligence agent. After exploring, you possibly can deploy to AWS and monitor your workflows within the AWS Administration Console.

The aggressive value intelligence agent is out there as an AWS Samples answer within the Amazon Nova Samples GitHub repository as a part of the Value Comparability use case.

1. Stipulations

Your improvement atmosphere should embody: Python: 3.10 or later and the Nova Act SDK.

2. Get Nova Act API key:

Navigate to https://nova.amazon.com/act and generate an API key. When utilizing the Nova Act Playground or selecting Nova Act developer instruments with API key authentication, entry and use are topic to the nova.amazon.com Phrases of Use.

3. Clone the repo, set the API key, and set up the dependencies:

To get began, clone the repository, set your API key so the applying can authenticate, and set up the required Python dependencies. This prepares your atmosphere so you possibly can run the mission regionally with out points. An API Key might be generated on Nova Act.

# Clone the repo 
https://github.com/aws-samples/amazon-nova-samples.git 
cd nova-act/usecases/price_comparison 

# Create and activate a digital atmosphere (non-compulsory however really helpful) 
python3 -m venv .venv 
supply .venv/bin/activate 

# Home windows:
.venvScriptsactivate 

# Set up Python dependencies 
pip set up -r necessities.txt 

# Set the Nova Act API Key export NOVA_ACT_API_KEY="your_api_key"

4. Operating the script

As soon as your atmosphere is ready up, you possibly can run the agent to carry out aggressive value intelligence. The script takes a product title (non-compulsory) and a listing of competitor web sites (non-compulsory), launches concurrent Nova Act browser classes, searches every website, extracts value and promotional particulars, and returns a structured, aggregated end result.

The earlier instance makes use of the script’s default competitor checklist, which incorporates main retailers resembling Amazon, Goal, Finest Purchase, and Costco. You possibly can override these defaults by supplying your personal checklist of competitor URLs when working the script.

python -m major.py 
    --product_name "iPad Professional 13-inch, 256GB Wi-Fi" 
    --product_sku "MVX23LL/A" 
    --headless

The agent launches a number of Nova Act browser classes in parallel, one per competitor website. Every session hundreds the retailer’s web site, checks whether or not a captcha is current, and pauses for person enter if one must be solved. As soon as clear, the agent searches for the product, evaluations the returned outcomes, clicks essentially the most related itemizing, and extracts the value and promotional data. Operating these flows concurrently permits the agent to finish a multi-site comparability effectively.

For instance, when focusing on Amazon, the agent opens a contemporary browser session, navigates to amazon.com, and performs a site-specific seek for the product. It inspects the returned outcomes, identifies the product itemizing that almost all carefully matches the question, and extracts key particulars resembling value, promotions, availability, and related metadata. This course of is mirrored within the following terminal output that displays every reasoning step (costs on this instance are illustrative and never consultant of actual market costs):

583c> act("Is there a captcha on the display?, format output with jsonschema: {"sort": "boolean"}")
583c> ...
583c> suppose("I'm on the Amazon homepage. My job is to return whether or not there's a captcha on the display. I can look across the web page to attempt to discover a captcha. I do not see something that appears like a captcha. I additionally do not see something that looks like it could require a captcha to be displayed. I ought to return false to point that there isn't a captcha on the web page.");
...
583c> act("Seek for 'iPad Professional 13-inch (M4 chip), 256GB Wi-Fi'.")
583c> ...
583c> suppose("I'm on the amazon homepage. My job is to seek for 'iPad Professional 13-inch (M4 chip), 256GB Wi-Fi'. I see an empty search discipline. I must sort 'iPad Professional 13-inch (M4 chip), 256GB Wi-Fi' into the search discipline.");
583c> ...
583c> suppose("I see that the search discipline is populated with 'iPad Professional 13-inch (M4 chip), 256GB Wi-Fi', however the search has not been initiated. Due to this fact, my final motion was not appropriate. I see a search button to the fitting of the search discipline which can be utilized to finish the search. I'll click on on the search button to finish the search.");
...
583c> suppose("I'm now on the search outcomes web page for 'iPad Professional 13-inch (M4 chip), 256GB Wi-Fi'. My job is full, and I must return.");
...
583c> suppose("I'm on the Amazon search outcomes web page for "iPad Professional 13-inch (M4 chip), 256GB Wi-Fi". My job is to assessment all the outcomes and discover essentially the most related one, then extract the value and any promotion particulars. I must return the knowledge within the specified JSON format. I see a listing of merchandise, however the first few are sponsored. I ought to scroll down the web page to search out essentially the most related, non-sponsored product.");
583c> ...
583c> suppose("The web page has scrolled down and I see extra merchandise. My prior motion of scrolling down was profitable. I see the primary non-sponsored product, which is the "Apple iPad Professional 13-Inch (M4): Constructed for Apple Intelligence, Extremely Retina XDR Show, 256GB, 12MP Entrance/Again Digicam, LiDAR Scanner, Wi-Fi 6E, Face ID, All-Day Battery Life — House Black". I ought to click on on the product to navigate to its web page and discover the value and promotion particulars.");
583c> ...
583c> suppose("I'm now on the product web page for the "Apple iPad Professional (M4) 13-inch (M4 chip), 256GB Wi-Fi". My prior motion of clicking on the product was profitable. I see the value of the product is $1,039.99 and there's a promotion for 19% off. I ought to return the value and promotion particulars within the specified JSON format.");

4. Reviewing the output

After the agent finishes looking all competitor websites, it returns a consolidated desk that lists every retailer, the matched product, the extracted value, the promotion particulars, and extra metadata. From this desk, you possibly can evaluate outcomes throughout a number of sources in a single view. For instance, the output would possibly look as follows (costs on this instance are illustrative and never consultant of actual market costs):

| Supply | Product Title | Product SKU | Value | Promotion Particulars |
|--------|--------------|-------|-------|-------------------|
| Amazon | Apple iPad Professional (M4) 13-inch (M4 chip), 256GB Wi-Fi | MVX23LL/A | $1,039.99 | 19% off |
| Finest Purchase | Apple - 13-inch iPad Professional M4 chip Constructed for Apple Intelligence Wi-Fi 256GB with OLED - Silver |  MVX23LL/A | $1239.00 | Save $50 |
| Costco | iPad Professional 13-inch (M4 chip), 256GB Wi-Fi | MVX23LL/A | $1039.99 | $200 OFF; financial savings is legitimate 11/12/25 via 11/22/25. Whereas provides final. Restrict 2 per member. |
| Goal | Apple iPad Professional (M4) WiFi with Customary glass | MVX23LL/A | $999.00 | Sale ends Wednesday |

The agent writes the extracted outcomes to a CSV file to later combine with pricing instruments, dashboards, or inner APIs.

Conclusion

Amazon Nova Act transforms browser automation from a fancy technical job right into a easy pure language interface, so retailers can automate guide workflows, cut back operational prices, and acquire real-time market insights. By considerably lowering the time spent on guide information assortment, groups can shift their focus to strategic pricing choices. The answer scales effectively as monitoring wants develop, with out requiring proportional will increase in sources. Nova Act allows builders to construct versatile, strong brokers that ship well timed insights, decrease operational effort, and help data-driven pricing choices throughout industries.

We welcome suggestions and would love to listen to how you employ Nova Act in your personal automation workflows. Share your ideas within the feedback part or open a dialogue within the GitHub repository. Go to the Nova Act to be taught extra or discover extra examples on the Amazon Nova Samples GitHub Repository.